1907.] Wheeler, The Polymorphism of Ants. 59 



selected, but the germ-plasm from which the body develops. The difference 

 is this : in the one case the survival in the struggle for existence depends on 

 characters and variations of the body of the individual; in the other, only 

 on the character of a certain kind of descendant — the worker. If the ant 

 state were composed of individuals connected together like a colony of poly- 

 pes or Sipho7iophorce, a process of selection by which only the workers were 

 changed would be within easier reach of our imagination, as these would 

 then, in a manner, be only organs, just like the snaring-threads, the swim- 

 ming bells, and the gastric tubes of the Siphonophoroe. As these do not 

 reproduce, and accordingly can only vary by selection of the egg or germ- 

 plasm from which the whole colony is formed, so in the case of the ant-colony, 

 or rather state, the barren individuals or organs are metamorphosed only by 

 selection of the germ-plasm from which the whole state proceeds. In respect 

 of selection the whole state behaves as a single animal ; the state is selected, 

 not the single individuals ; and the various forms behave like the parts of one 

 individual in the course of ordinary selection." 



It must be admitted that this hypothesis is boldly and clearly conceived 

 and its author's knowledge of myrmecology and melittology is only surpassed 

 by the adroitness with which he compells the facts to tally with his assump- 

 tions. Nowhere in Weismann's work are both the strength and the weak- 

 ness of his elaborate architectonics of the germ-plasm more apparent than in 

 this attempt to explain the complicated and adaptively correlated organi- 

 zation of the sterile worker as the result of a single mechanical factor, natural 

 selection, acting on constellations of fortuitous determinants in the germ- 

 plasm of the fertile female. 



Spencer, too, seeks a mechanical explanation of the purposeful adapta- 

 tions of the worker, though this is utterly at variance with the fundamental 

 teleological spirit of the Lamarckian view which he elsewhere advocates. 

 He starts out with the social wasps (Vespa), some species of which present 

 an uninterrupted series of transitional forms between the small sterile 

 worker and the large fertile female, or queen. He calls attention to the 

 abundant evidence to show that these arise progressively and in close depend- 

 ence on the food supply of the colony, so that the sterility and dwarfing of 

 the workers are manifestly caused by insufficiency of food during their larval 

 stages. He then proceeds to extend this same explanation to the ants, 

 although these insects differ from the wasps in presenting great morphologi- 

 cal differences between the workers and females. After considering the 

 African driver ants which have highly polymorphic workers, he goes on to 

 say: "In contrast with this interpretation, consider now that of Professor 

 Weismann. From whichever of the two possible suppositions he sets out, 

 the result is equally fatal. If he is consistent, he must say that each of these 



